Apparatus and method for improved digit recognition and caller identification in telephone mail messaging

• US 6,219,407 B1
• Filed: 01/16/1998
• Issued: 04/17/2001
• Est. Priority Date: 01/16/1998
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Kanevsky, Dimitri, Maes, Stephane Herman
• Primary Examiner(s): Tsang, Fan
• Assistant Examiner(s): Hoosain, Allan
• Application Number:
  US09/008,471
• Time in Patent Office: 1,187 Days
• Field of Search: 379/67.1, 379/71, 379/76, 379/80, 379/88.01, 379/88.02, 379/88.04, 379/88.07, 379/88.08, 379/88.22, 379/88.26, 379/88.27, 379/93.03, 379/157, 379/201, 379/93.23, 379/243, 704/243, 704/255, 704/260, 704/275
• US Class Current: 379/88.02
• CPC Class Codes: H04M 1/575 Means for retrieving and displaying personal data about calling party
• CPC Class Codes: H04M 1/6505 storing speech in digital form
• CPC Class Codes: H04M 11/066 Telephone sets adapted for data transmision

Abstract

A telephone voice mail messaging system for performing recognition of characters included in a telephone message left by a caller includes: speaker recognition means, responsive to the telephone message, for extracting a voice model of the caller from the telephone message and comparing the voice model of the caller against a plurality of pre-obtained voice model'"'"'s respectively associated with a plurality of potential callers to find a pre-obtained voice model from among the plurality of pre-obtained voice models that matches the voice model of the caller and, if the matching pre-obtained voice model is found, then the speaker recognition means identifying at least one pre-obtained set of characters attributed to a potential caller from among the plurality of potential callers associated with the matching pre-obtained voice model; character spotting means, responsive to the telephone message, for identifying sets of characters in the telephone message; and comparator means for comparing characters of the at least one pre-obtained set of characters identified by the speaker recognition means against characters of a set of characters identified by the character spotting means and, if the characters match, then outputting the characters of the at least one pre-obtained set of characters as representing the set of characters left by the caller as part of the telephone message.

Claims

1. A telephone voice mail messaging system for performing recognition of characters included in a telephone message left by a caller, the system comprising:

speaker recognition means, responsive to the telephone message, for extracting a voice model of the caller from the telephone message and comparing the voice model of the caller against a plurality of pre-obtained voice models respectively associated with a plurality of potential callers to find a pre-obtained voice model from among the plurality of pre-obtained voice models that matches the voice model of the caller and, if the matching pre-obtained voice model is found, then the speaker recognition means identifying at least one pre-obtained set of characters attributed to a potential caller from among the plurality of potential callers associated with the matching pre-obtained voice model;

character spotting means, responsive to the telephone message, for identifying sets of characters in the telephone message; and

comparator means for comparing characters of the at least one pre-obtained set of characters identified by the speaker recognition means against characters of a set of characters identified by the character spotting means and, if the characters match, then outputting the characters of the at least one pre-obtained set of characters as representing the set of characters left by the caller as part of the telephone message.

2. The system of claim 1, further comprising message recognition means, response to the telephone message, for decoding the telephone message into a decoded message representative of the telephone message which may include one of a name of the caller, a telephone number of the caller and a statement made by the caller.

3. The system of claim 2, wherein the message recognition means further comprises means for decoding the telephone message using one of a class-based model and speaker trained model identified through the speaker recognition means.

4. The system of claim 2, further comprising message correction means, responsive to the set of characters output by the comparator means and the decoded message output by the message recognition means, for replacing a corresponding set of characters decoded by the message recognition means with the set of characters output by the comparator means to form a corrected decoded message.

5. The system of claim 4, further comprising user interface means, responsive to the corrected decoded message, for displaying to a user one of the name, the telephone number, and the statement of the caller from the corrected decoded message.

6. The system of claim 5, wherein the system is utilized as a user paging system and the user interface unit is a pager device.

7. The system of claim 5, wherein the system further comprises:

telephonic means, responsive to the user interface means, for permitting the user to place a telephone call; and

selection means, operatively coupled to the user interface means, for permitting the user to select the telephone number of the caller and thus initiate a return call to the caller via the telephonic means.

8. The system of claim 5, wherein the user interface means further comprises means for selectively displaying the statement to the user.

9. The system of claim 5, wherein the user interface means further comprises means for selectively initiating an audio playback of the telephone message to the user.

10. The system of claim 1, wherein the set of characters identified by the character spotting means are limited to the characters of the at least one pre-obtained set of characters identified by the speaker recognition means.

11. The system of claim 1, wherein the pre-obtained set of characters attributed to the potential callers and the set of characters identified by the character spotting means are respectively one of a name, a telephone number and an address.

12. The system of claim 1, further comprising user interface means, responsive to the comparator means, for displaying to the user one of the set of characters identified by the character spotting means and the at least one pre-obtained set of characters from the speaker recognition means.

13. The system of claim 12, wherein the sets of characters are telephone numbers and wherein the system further comprises:

telephonic means, responsive to the user interface means, for permitting the user to place a telephone call; and

selection means, operatively coupled to the user interface means, for permitting the user to select the telephone number of the caller and thus initiate a return call to the caller via the telephonic means.

14. The system of claim 1, further comprising means for enrolling the caller in the system if a substantial match is not found by the speaker recognition means.

15. A telephone voice mail messaging system for performing caller identification with respect to a telephone message left by a caller, the system comprising:

character spotting means, responsive to the telephone message, for identifying sets of characters in the telephone message and comparing an identified set of characters against a plurality of pre-obtained sets of characters associated with a plurality of potential callers to find a pre-obtained set of characters from among the plurality of pre-obtained sets of characters that matches the identified set of characters and, if the matching pre-obtained set of characters is found, then the character spotting means further identifying a list of potential callers attributed to the matching pre-obtained set of characters; and

speaker identification means, responsive to the telephone message, for extracting a voice model of the caller from the telephone message and comparing the voice model of the caller against respective voice models associated with potential callers on the list to find a respective voice model from among the respective voice models that matches the voice model of the caller and, if the matching respective voice model is found, then the speaker identification means identifying a potential caller from the list associated with the matching respective voice model.

16. The system of claim 15, wherein the pre-obtained set of characters and the set of characters identified by the character spotting means include a telephone number.

17. The system of claim 16, further comprising means for identifying an actual telephone number associated with the identified caller.

18. The system of claim 17, further comprising user interface means, responsive to the character spotting means, for displaying to a user one of a name associated with the identified caller and the telephone number associated with the identified caller.

19. The system of claim 18, wherein the system is utilized as a user paging system and the user interface unit is a pager device.

20. The system of claim 15, further comprising message recognition means, responsive to the telephone message, for decoding the telephone message into a decoded message representative of the telephone message which may include a statement made by the caller.

21. The system of claim 20, wherein the message recognition means further comprises means for decoding the telephone message using one of a class-based model and speaker trained model identified through the speaker identification means.

22. The system of claim 20, further comprising user interface means, responsive to the character spotting means, for displaying to a user the statement of the caller.

23. The system of claim 22, wherein the user interface means further comprises means for selectively displaying the statement to the user.

24. The system of claim 22, wherein the user interface means further comprises means for selectively initiating an audio playback of the telephone message to the user.

25. The system of claim 15, further comprising means for selectively permitting a caller access to a service in response to the identity of the caller.

26. The system of claim 15, further comprising means for enrolling the caller in the system if a substantial match is not found by the speaker identification means.

27. A method for performing recognition of characters included in a telephone message left by a caller in a telephone voice mail messaging system, the method comprising the steps of:

(a) extracting a voice model of the caller from the telephone message;

(b) comparing the voice model of the caller against a plurality of pre-obtained voice models respectively associated with a plurality of potential callers to find a pre-obtained voice model from among the plurality of pre-obtained voice models that matches the voice model of the caller;

(c) if the matching pre-obtained voice model is found, then identifying at least one pre-obtained set of characters attributed to a potential caller from among the plurality of potential callers associated with the matching pre-obtained voice model;

(d) identifying sets of characters in the telephone message;

(e) comparing characters of the at least one pre-obtained set of characters identified in step (c) against characters of a set of characters identified in step (d); and

(f) if the characters match, then outputting the characters of the at least one pre-obtained set of characters as representing the set of characters left by the caller as part of the telephone message.

28. The method of claim 27, further comprising the step of decoding the telephone message into a decoded message representative of the telephone message which may include one of a name of the caller, a telephone number of the caller and a statement made by the caller.

29. The method of claim 28, wherein the decoding step further comprises decoding the telephone message using one of a class-based model and a speaker trained model identified via step (b).

30. The method of claim 28, further comprising the step of replacing a set of corresponding characters in the decoded message with the set of characters output in step (f) to form a corrected decoded message.

31. The method of claim 30, further comprising the step of displaying to a user one of the name, the telephone number, and the statement of the caller from the corrected decoded message.

32. The method of claim 31, wherein the displaying step further comprises displaying one of the name, the telephone number and the statement of the caller to the user on a paging device.

33. The method of claim 27, further comprising the step of enrolling the caller in the system if a substantial match is not found via step (b).

34. A method for performing caller identification with respect to a telephone message left by a caller in a telephone voice mail messaging system, the method comprising the steps of:

(a) identifying sets of characters in the telephone message;

(b) comparing an identified set of characters against a plurality of pre-obtained sets of characters associated with a plurality of potential callers to find a a pre-obtained set of characters from among the plurality of pre-obtained sets of characters that matches the identified set of characters;

(c) if the matching pre-obtained set of characters is found, then identifying a list of potential callers attributed to the matching pre-obtained set of characters;

(d) extracting a voice model of the caller from the telephone message;

(e) comparing the voice model of the caller against respective voice models associated with potential callers on the list to find a respective voice model from among the respective voice models that matches the voice model of the caller; and

(f) if the matching respective voice model is found, then identifying a potential caller from the list associated with the matching respective voice model.

35. The method of claim 34, wherein the pre-obtained set of characters and the set of characters identified in step (a) include a telephone number.

36. The method of claim 35, further comprising the step of identifying an actual telephone number associated with the identified caller.

37. The method of claim 36, further comprising the step of displaying to a user one of a name associated with the identified caller and the telephone number associated with the identified caller.

38. The method of claim 37, wherein the displaying step further comprises displaying one of the name and the telephone number of the caller to the user on a paging device.

39. The method of claim 34, further comprising the step of decoding the telephone message into a decoded message representative of the telephone message which may include a statement made by the caller.

40. The method of claim 39, wherein the decoding step further comprises decoding the telephone message using one of a class-based model and a speaker trained model identified via step (e).

41. The method of claim 39, further comprising the step of displaying to a user the statement of the caller.

42. The method of claim 34, further comprising the step of enrolling the caller in the system if a substantial match is not found via step (e).